I have been working on a nav module for RC models in order that they drive themselves. My first application is boats because it gets over difficulties such as obstcles and auto-flight stabalisation etc... I think this is an ideal add-on to a boat in order to get it back to the water's edge when it goes out of range. It also allows you to see just what your boat can do without the limitations of remote control range.

The basic principal is that you specify an end coordinate (you can get that from the on board GPS or even google maps etc...) and then tell it the waypoints you want it to pick up along the way (again from a GPS or map). It will then follow the waypoints and end up at the end point you specify.

The cool thing is that you can watch your boat go flat out over some distance and have it return to you once it has reached a point in the distance. Also if you end up with it going out of range of your transmitter you can set a, say 1 minute, stationary timeout and have it return to your end point on its own using GPS.

The cost in parts is about £100 maybe $150. I am happy to share the firmware I have written at this stage with modellers so that they can evaluate the concept. I suppose there is the potential for a commercial product if one doesn't already exist but I am just interested in it working at the moment.

What do you guys (and gals!) think?

For some this will be a great idea... for those that cant control their model, of course...

I thought the idea of RC stuff was so you could control it yourself?...

Ofcourse it is.

I think you are right about it being good for people who 'cant control their model' but for experienced modellers it can allow you to sail closer to the wind if you dont mind the pun. Like traction control on race cars it just allows you to get more out of what you have without worrying about it going wrong.

As far as watching it sail off on its own then I suppose this would be fun for a few minutes and then get a bit tiresome. However if you like tuning engines and making the thing as fast as possible then having an on-board controller than can sail it flat out in a straight line to test it's speed and record it's top speed is the best way to see what your machine can really do.

Actually, if the module could navigate the way points, sail to a dock, reach out and load an object, then
navigate the same waypoints back and depost the object in a goal area, that would be useful.
I know some high schoolers that could use that kind of programmability.

[8D]

Being a high school teacher I coulden't agree more!

It could but GPS is only accurate to 5m (realistically +-3m is possible). Once it is within range of a loading bay then something else would be required for precise positioning. Probably an optical positioning system is best but there may be some acoustic/ultrasonic system which would work.

Similarly it needs to be doing at least 5mph to get an accurate heading reference either from the GPS module or from the GPS generated difference in latitude/longtitude co-ords. The idea is to calculate a bearing to the desired co-ords from the current co-ords and follow a line on this heading. The steering would be performed by multiplying the difference in heading and bearing by some gain value (proportional control).

On my design I have used an 8-bit processor and I calculate ATAN and SIN/COS with some 24-bit maths subroutines I have written. I am happy to share this code if anyone would like to see it. It is written for PIC micros which I believe is taught in schools these days.

Maybe you could design a system to precisely position a boat in school.

5m?... [X(]
I must not crash the dock.. I must not crash the dock...

The idea isn't new for full sized boats, but for R/C sized boats it certainly would be. Biggest problem is miniaturizing the whole thing, not to mention financing it. I think that 100 - 150 Pound range is -VERY- optimistic! The accuracy of such a GPS based navigation system is 'workable' for 'full sized' things. That accuracy is almost 5 times(?) the size of a typical R/C boat. That might get the boat back to a bank, but I don't think I'd count on it being near your feet, on that bank (unless you plan to do a lot of walking, sort of). Now if you can 'scale' the accuracy up as much as the R/C boat is 'scaled' down... Oh have I got a deal for you!!
- 'Doc

Pickey, picky.

You are no fun.

Well you can use differential GPS to get that accuracy up to better than 1m. There are probably land based beacons by a lot of large bodies of water to provide an additional GPS source.

Still by the bank is better than a kilometer away!

Cost wise? GPS modules are about £30-£40. I am using a mere 8 bit processor in my prototype and have yet to max it out. This should leave the basic navigation system done for around the £50 mark. However I think if it were a commercial system you would need to add some gyro sensors to provide an additional approximation for direction whilst steering. They are expensive.

On that DGPS subject i decided to look for my local beacon and it really is rather close. I imagine that there are lots around where you yourself live. They need to be within 100km or so.

http://www.trimble.com/findbeacon.asp

While I hope your GPS idea does work, I'm afraid that the typical R/C'er might find that it isn't too practical for use. Cost is one of those 'impractical' things, something in the neighborhood of $300 - $400 just isn't a normal part of my R/C'ing budget to get a boat back to the bank...without financing by that bank (pun intended).
Accuracy is still not scaled up enough even with a shore beacon. Nearest one to me is a bit over 100 miles away. Considering the terrain between it and me, and the typical behavior of UHF signals, the GPS I use locally has no idea the beacon even exists.
- 'Doc

Off topic part.

I 'play' with a system that uses GPS for position reporting called APRS (connected to another expensive hobby I have). I track my position using a computer and a GPS receiver to furnish information. It's accurate 'enough'. Which, considering the accuracy of the mapping program and the 'size' of the icon representing my vehicle, will put the reported position within the tolerances of GPS accuracy. This thingy does about the same thing that the 'Delorme' GPS tracking program ('Street Atlas') does, doesn't provide the same statistics, report options though. But, I (and no telling how many others also) put the mess together and made it work. It requires a beacon in the tracked vehicle, a radio receiver to pick up that beacons signal, and a computer to 'mix' the whole thing and give you a picture of where the beacon is. Nothing particularly new about it, and it's done commercially too. I can see applications for model boating, but am just too cheap to try it. Why go to all this trouble when there are commercially available systems to do the same thing? Cuz I'm one of those 'radio freaks', you know?
Oh, the programming was done by one of you 'UK' guys. I'm sorry to say he isn't with us anymore.

Well for the moment it is just fun but the target price I would expect would be more like £50 ($80ish?).


It probably would be more accurate to use two beacons at the bank to navigate by but maybe less user friendly.

There are quite a number of 'possible' solutions to getting a boat back to -a- bank. Getting it back to a specific point on -a- bank complicates things by several orders of magnitude.
If this system works when there is a loss of signal, would it 'kick-in' when ever you stop the boat still. Distinguishing between intentional and accidental loss of signal would certainly be nice. There's no constant signal with the typical R/C radio system, it depends on the operator issuing a control movement, it's intermittent. Redesigning R/C radio systems isn't very practical an idea for a number of reasons (battery size?). A waiting period before activation? I can see problems with that for IC powered boats and 'fail-safes', how do you restart that @#$ IC engine?

Please don't misunderstand, I think you have a pretty nice idea. I honestly don't think it's too practical right now, and for R/C'ing, but it certainly does have applications! I wish you luck with it.
I'd drop hints about you letting me do the marketing for it, but you ain't that dumb to let me, so I won't...RATS!
- 'Doc